Application to manage Fraud Transactions based on Spring Boot Rest Api. Provides a processing Transaction Files with Spring Batch and expose an Api to trigger job.

• Lightweight Spring Boot service with embedded tomcat and use Spring Batch to proceeding transaction list files.
• Rest Api is documented with Swagger and openApi, which provides nice UI with all documentation and ability to test services
• Persistence based on Spring Data JPA and in-memory h2 database to simplify the deployment and avoid external dependencies. (not suitable for production, but easily replaceable by other relational database)
• Unit and integration testing with JUnit and Mockito. Rest Api integration testing with spring-test.

To be able to run application you need to have installed:

• Spring webmvc/jpa data/hibernate
• Maven
• Java 17
• h2
• Docker
• Swagger/OpenApi/Springfox
• Mockito

By default application will run in development mode (see application.properties):

• Logs are printed only to console and root level for application packages - DEBUG
• Application uses in-memory database

To run in production mode, application should be started with spring profile "server" activated (see application-server.properties):

• Logs will be printed to file in the logs folder of the root directory
• Logging root level for all packages - INFO

Run the docker compose file

      docker-compose -f "docker-compose.yml" up --build -d

With maven:
Use spring boot plugin to start(it will build backend and frontend):
mvn spring-boot:run

For deployment on test/production without docker server:
Build jar package with maven:
mvn clean package
Run application with:
java -jar fraud-detection-0.0.1-SNAPSHOT.jar --spring.profiles.active=server

Example of request to trigger the job for processing CSV file and return fraud list:

curl --location --request GET 'http://localhost:9595/job/CSV'

Example of request to trigger the job for processing JSON file and return fraud list:

curl --location --request GET 'http://localhost:9595/job/JSON'

Or you can use Swagger UI, which contains Rest Api documentation:
http://localhost:9595/swagger-ui/index.html

• Application uses in-memory relational database h2, so after restart all the data is reloaded.
• On startup application creates tables in the database based on JPA-entities, which is not suitable for production.

• Move to proper persistence storage: relational or NoSql database.
• Use encrypted transport protocol(TLS) for Rest Api.
• Improve authentication/authorization with proper user/roles management.
• Add support for encrypted properties to avoid sensitive data provided in plain text.
• Provide some service to upload and manage the Fraud file with different formats
• Add a AMQP like Kafka to manage and detect fraud detection in real time.
• Configure CI/CD server for build and deployment
• Improve input validation for Rest Api. Define better error messages in property file.
• Add pagination for GET all frauds api to be able to limit amount of returned items.
• Add separate maven profile to run integration tests.